|
Qarshi davlat universiteti international scientific and practical conference on algorithms and current problems of programming
Pdf ko'rish
|
| bet | 38/551 | | Sana | 15.05.2024 | | Hajmi | 15,84 Mb. | | #234763 |
Bog'liq Asosiy oxirgi 17.05.2023 18.20
LITERATURE ANALYSIS
: A comprehensive literature analysis was conducted to
understand the current state of research in Braille text recognition and edge-based filtering
techniques. Previous studies have employed various approaches, such as template
matching, machine learning, and image processing techniques, for Braille text recognition.
However, these methods may face challenges in accurately detecting the Braille dots, which
are crucial for recognition.
Edge-based filtering techniques, such as Canny, Sobel, and Scharr, have been used in
other image processing tasks, such as edge detection, object extraction, and region of
interest detection. These techniques have shown promising results in enhancing the
accuracy of these tasks, which makes them potential candidates for improving Braille text
recognition as well.
METHOD: The proposed method utilizes edge-based filtering techniques, including
Canny, Sobel, and Scharr, for enhancing Braille text recognition. The thematic formulas and
graphs presented in this section demonstrate the effectiveness of the approach[4]. The
images containing Braille text are preprocessed to remove noise and normalize the
intensity levels [5]. Then, the edge-based filtering techniques are applied to detect the
edges of the Braille dots. The filtered edge maps are further processed to extract the Braille
text regions [6]. Finally, the extracted regions are fed into a Braille text recognition
algorithm for accurate recognition.
RESULTS: To evaluate the effectiveness of the proposed method, a comprehensive
experimental evaluation was conducted on a dataset of images containing Braille text. The
results are presented in an accurate calculation table, showcasing the performance of the
edge-based filtering techniques in terms of accuracy, precision, recall, and F1 score [7-8].
The results demonstrate that the proposed method using edge-based filtering techniques
significantly improves the extraction accuracy of Braille text compared to existing methods
[9].
CONCLUSION
In this article, we have discussed how edge-based filtering techniques, such as Canny,
Sobel, and Scharr, can be utilized to enhance the extraction of Braille text from images,
leading to improved accuracy in Braille text recognition [10]. The presented method, along
with the thematic formulas, graphs, and calculation table, showcases the effectiveness of
the approach.
52
The experimental results validate the potential of edge-based filtering techniques for
enhancing Braille text recognition and highlight their significance in enabling visually
impaired individuals to access written information [11]. Further research can focus on
optimizing the proposed method for real-world applications and exploring additional
techniques for improving Braille text recognition accuracy.
|
|
Bosh sahifa
Aloqalar
Bosh sahifa
Qarshi davlat universiteti international scientific and practical conference on algorithms and current problems of programming
|
